Device and method for generating label objects for the surroundings of a vehicle

Abstract

A method and a labeling system for generating a label object for the symbolic description of an object of an environment of a mobile device, e.g., a robot or a vehicle. The label object includes at least one attribute of an object at a first point in time, from observations of this object. The method includes selecting, from the observations, a first observation recorded at a first point in time, a second observation recorded at a second point in time, the second point in time being a point in time before the first point in time, as well as a third observation recorded at a third point in time, the third point in time being a point in time after the first point in time; and ascertaining, by using the selected observations, the at least one attribute of the object.

Description

CROSS REFERENCE[0001]The present application claims the benefit under 35 U.S.C. § 119 of German Patent Application No. DE 102018220892.1 filed on Dec. 4, 2018, which is expressly incorporated herein by reference in its entirety.FIELD[0002]The present invention relates to a method and a labeling system for generating a label object for the symbolic description of an object of an environment, or a surroundings, of a mobile device, e.g., a robot or a vehicle, in particular of a mobile device that is movable at least partially in automated fashion. The present invention furthermore relates to a program element, a computer-readable medium and a use.BACKGROUND INFORMATION[0003]For a mobile device, e.g., for a robot or a vehicle, a symbolic description of an object, or a plurality of objects, of an environment or a surroundings of the mobile device may be important. For the symbolic description of the object, the object is recorded, e.g., by a camera and is provided with attributes, e.g., ...

AI Technical Summary

Benefits of technology

The patent describes a system for labeling objects using sensor data collected from a vehicle. The system uses a combination of pattern detection and multi-target tracking algorithms to identify attributes of the objects. The labels are generated using a holistic processing of the observation, which takes into account the present, past, and future of the observation. The system can also use reference sensors, such as calibrated sensors or additional sensors, to improve the accuracy of the labels. The classification of the objects can be performed using an artificial neural network or a Bayesian filter. The system can store the sensor data on a persistent storage medium and transfer it to a server computer or cluster for further analysis and labeling. Overall, the system achieves a higher accuracy and reliability of the labels and allows for the use of more advanced algorithms and machine learning methods.

Problems solved by technology

The detected and “labeled” object may be called a “label object.” These manual methods may be expensive in terms of time and costs, however.

In one specific embodiment (e.g., of offline processing), the iterative process does not have to fulfill real time conditions and may therefore be comparatively time-intensive and computation-intensive.

Images